Redistributive politics in post-independence fiji - Core
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Pacific Economic Bulletin
50/50 by 2020: poverty and
redistributive politics in
post-independence Fiji
Satish Chand
The affirmative action program launched by the Fiji Government in Satish Chand is
2002 espoused a ‘50/50 by 2020’ vision; that is, by the year 2020, some Associate Professor
50 per cent of all economic activities would be owned by the indigenous and Director of the
Pacific Policy Project
population. The surprising impact of this heavy-handed redistribution of
in the Crawford
income and wealth from the non-indigenous to the indigenous population School of Economics
has been poverty-raising. One in eight in the population lived in poverty and Government
in 1977, the figure had risen to one in four by 1990/91 and one in three at The Australian
by 2002/2003; and on current trends, would reach one in two (that is 50 National University.
per cent) by 2020. Such an outcome would be a direct consequence of
these redistributive policies. While the politics of redistribution may have
been compelling, its economic costs, including the impact on poverty, are
devastating. The 2006 military takeover was executed to rid the country
of corruption and race-based politics. Achieving these goals may have a
bonus in terms of reversing the rise in poverty. Only time will tell.
Poverty in Fiji has been increasing over at birth rose from 60.6 years in 1970 to 67.8
time. 1 Both the incidence (that is the years by 2004, per capita GDP has grown at
proportion of people living in poverty) and an annual average rate of one per cent since
the severity of poverty (that is, the difference 1975, reaching a purchasing power parity
between the poverty line and the mean (PPP) figure of F$6,066 by 2004, and the adult
income of the poor) increased between the literacy rate as of 2003 was 93 per cent as
three Household Income and Expenditure compared to 88.6 per cent in 1990. The HDI, an
Surveys (HIES); those of 1977, 1990/91, and aggregate of these variables, climbed steadily
2002/03. The rise in poverty, however, was from 0.663 in 1975 to 0.758 by 2004 (UNDP
accompanied by an improvement in the 2006). The simultaneous rise in the HDI and
Human Development Index (HDI) for the the incidence of poverty suggests two things:
nation as a whole. Average life expectancy first that the benefits of development have
22
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
been unevenly distributed; and, second that is healthy for an economy as it is the largest
the poor have lost ground, possibly falling source of tax revenues, comprises the most
into deepened pockets of disadvantage. The productive of the workforce, and is often a
rise in poverty accompanying modest growth source of demand for improved governance.
in the economy is evidence of failure of the The loss of this group, therefore, is to the cost
many pro-poor policy pronouncements of of each of the above-mentioned. While some
the past.2 level of emigration has been taking place
The rise in poverty is likely to continue since the 1960s, the levels of emigration
given a slowing economy, and the eviction of nearly doubled following the first coup of
many rural farmers from land with expired 1987 and with spikes in the numbers leaving
leases. Government-sponsored redistribution after each of the next two coups (Chandra
has been based on ethnicity under the (false) and Chetty 1996; Asia Pacific Migration
premise that only the indigenous population Research Network n.d.). The coups no
is poor. Furthermore, the affirmative action doubt have been responsible for hollowing
polices have lent themselves to abuse such out the middle of the economy and thus
that the redistribution has driven a wedge contributing to the rising trend in poverty
between the rich and the poor in both of the of the past two decades.
major ethnic groups. Indigenous Fijians who
have been able to access public handouts Poverty viewed via the ethnic lens
were anything but poor. These transfers
have been at the expense of the poorest Indo- Contrary to the claims of the recently ousted
Fijians who lack the capacity to fund goods government in justifying affirmative action
and services that are provided from the programs for the indigenous population,
public purse to the indigenous population. poverty pervades all communities in
The politics of redistribution suits the élite— Fiji. Importantly, there is no discernable
the indigenous Fijian élite in particular—as statistical difference in the levels of poverty
it legitimises the transfers to them but is between Indo-Fijians and indigenous Fijians.
couched as helping the ‘poor Fijians’. The Redistributions based on the ethnicity of
Indo-Fijian élite, many with long-established the recipient, therefore, are likely to be
links to the government, and/or the means ineffective and, as argued in the next section,
to circumvent (and even benefit from) the even counterproductive.
discriminatory policies have little to lose Before presenting evidence on poverty
from affirmative action programs. levels, let me define (absolute) poverty. A
The middle class, constituting mostly household or individual is considered to
professional and trade personnel, comprise be in absolute poverty when they lack the
the majority of the taxpayers and thus have income to afford the minimal bundle of food
funded most of the public handouts. The and shelter, that is, the basic necessities of
mobile amongst this group, not surprisingly, life, for healthy living. Food poverty refers
have chosen to seek opportunities abroad. to the situation where the individual lacks
This exodus, while initially comprising the means to acquire the minimal dietary
Indo-Fijians, has now spilled across to requirements for healthy living. Food
indigenous Fijians with some of the loss poverty in the form of abject destitution that
of human capital being compensated for is visible in parts of Asia and Africa is not
via rising receipts of remittance income. As yet evident in Fiji, but there are many within
of 2006, remittance receipts had reached the community who live hand-to-mouth
some F$400 million. A middle-class bulge and thus are highly vulnerable to falling
23
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressPacific Economic Bulletin
into food-poverty. Furthermore, social then, some 11.6 per cent of the population
welfare payments from the government was in poverty in 1977. Using the same
have provided only a limited safety net to methodology but now with an appropriately
the most disadvantaged in society. constructed poverty line, some 25.5 per cent
Statistics on the level of poverty have of the population was in poverty with
been compiled from analysis of household some 10 per cent of the households in food-
level data, collected via the comprehensive poverty in 1990/91 when the next HIES
HIES of 1977, 1990/91, and 2002/03. The was conducted (UNDP 1996:33). In the 14
compilation of poverty statistics involves years between the two surveys of 1977
computing the poverty line, counting the and 1990/91, poverty rose by an annual
number of people below this line to reach average of one percentage point a year. As
the headcount measure, and working out regards the ethnic composition of those
the gap between the average income of in poverty, some 31 per cent of the Indian
those in poverty and the poverty line as households had incomes below the poverty
the measure of the severity of poverty. line while the corresponding figure for Fijian
The poverty line establishes the minimum households was 27.6 per cent. Furthermore,
income a household needs to provide for the flat income distribution around the
its basic needs; it thus takes into account poverty line in 1990/91 implied that a large
the composition of the consumption basket proportion of households were vulnerable
as well as the price and quantities of each to falling into poverty with a deterioration
of the items within this basket. It is only in economic conditions.
a threshold in terms of the minimum The absence of detailed analysis of
purchasing power necessary to fund the poverty by ethnicity in 1977 precludes
minimal requirements for healthy living; comparisons with the 1990/91 data in terms
however, it makes no assumption regarding of any changes in the ethnic mix of the poor,
how the income is expended. An income less but these comparisons can be made between
than the poverty line is interpreted as the the HIES of 1990/91 and 2002/03. However,
particular household being in poverty. one comparison, that of average daily wages
In contrast to the concept of absolute between 1977 and 1990 is suggestive of
poverty defined above, relative poverty (that the deteriorating economic conditions for
is, income inequality) refers to the situation those who drew their income from the sale
where a part of the population has a much of labour alone. The average real (that is,
smaller share of national income than inflation adjusted) daily wage rate of 1990
the rest of the population.3 However, this was only 62 per cent of the corresponding
says nothing about whether these people figure for 1970 (UNDP 1994). While per
are in absolute poverty. As an example, a capita GDP grew by an average of one per
household with an income in the bottom cent per annum in the two decades following
20 per cent of the income distribution may independence, wages fell by approximately
be considered relatively poor but this says 2 per cent a year in this period.
nothing about their capacity, or otherwise, The Fiji Poverty Report, produced by the
of acquiring the minimal necessities for UNDP together with the government (UNDP
healthy living. For the purposes of this 1997) drew the attention of policymakers
article, unless otherwise stated, poverty will to the plight of the poor and the need to
mean absolute poverty. reverse the worsening trends in poverty. The
Based on the diets of the population government announced its determination to
in 1977 and the costs of basic necessities reduce poverty from the levels observed in
24
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
1990/91; it announced that sustained growth T h i s c o n c l u s i o n re g a rd i n g t h e
of the economy was the key instrument to pervasiveness of poverty across the major
be used to enable the disadvantaged to enter two ethnic groups is reaffirmed by the
the workforce and lift themselves out of more recent analysis of the 2002/03 HIES
poverty (Fiji 1993, 1996b). Jobs growth was data. Abbott (2007) notes that ‘[a]mongst
to be the central plank of the strategy. In Fijian households the proportion having
addition, the public sector was to concentrate per capita incomes below the BNPL [basic
its efforts on providing basic services and the needs poverty line] is estimated at 29.0 per
regulatory conditions for private enterprise. cent (33.8 per cent of the Fijian population)
However, the record on each of these goals and 30.7 per cent for Indian households
is less than enviable. (34.9 per cent of the Indian population)’
Poverty is pervasive across the two (Abbott 2007:22).
major ethnic groups. The Fiji Poverty Report, The incidence of poverty across the
for example, notes that ‘[t]he data on poor two major ethnic groups is not too different
households…revealed that poverty is between the HIES of 2002/03 and that of
not concentrated in one particular sector, 1990/91. In terms of basic needs, some
being neither predominantly rural or 33.8 per cent of the Fijian population
urban, nor exclusive to one ethnic group, was in poverty in 2002/03, an increase
but pervades all communities and areas’ from 27.7 per cent for 1990/91, while the
(UNDP 1997:103). According to the report, corresponding figure for Indo-Fijians was
the poorest in Fiji were those who earned 34.9 per cent and 31.0 per cent in 2002/03
income from doing menial (unskilled) jobs, and 1990/91, respectively.
had poor access to land or had little security According to the HIES of 2002/03, some
to the land they occupied, and had no other 38.1 per cent of the rural population and
sources of income. 31.8 per cent of the urban population was
Table 1 Incidence of poverty, 1990/91 and 2002/03 (per cent of population)
Per cent of population having an adult equivalent
per capita income less than the
Basic Needs Poverty Line
1990/91 2002
National average 25.5 34.4
Indigenous Fijian 27.7 33.8
Indo-Fijian 31.0 34.9
Urban average 27.6 31.8
Indigenous Fijian .. 29.9
Indo-Fijian .. 33.9
Rural average 24.3 38.1
Indigenous Fijian .. 37.3
Indo-Fijian .. 39.2
Source: Abbott, D.F., 2007. Fiji analysis of the 2002/03 household income and expenditure surveys: estimation
of basic needs poverty lines and incidence of poverty in Fiji, Draft Final Report, prepared for United Nations
Development Programme, Suva (unpublished):23.
25
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressPacific Economic Bulletin
in poverty compared to the corresponding Income inequality amongst Indo-Fijians
figures of 27.6 per cent for the urban is greater than that amongst indigenous
population and 24.3 per cent for the rural Fijians. According to income figures from
population as of the 1990/91 HIES. While the HIES of 1990/91, indigenous Fijian
the 1990/91 HIES data does not disaggregate households had an average weekly income
rural and urban poverty by ethnicity, the of F$173.65, some 13 per cent lower than
2002/03 HIES data shows that some 33.9 the national average of F$199.31; Indo-
per cent of urban Indo-Fijians and 29.9 per Fijian households had an average income
cent of urban indigenous Fijians were in of F$217.89 (some 9 per cent above the
poverty, while the corresponding figures for national average); and, ‘Other’ households
the rural population was 39.2 per cent for had an average income of F$271.08 (some 36
Indo-Fijians and 37.3 per cent for indigenous per cent above the national average). While
Fijians (UNDP 2006:3). Indo-Fijian households had a higher average
As of 2002/03, Indo-Fijian households income than their Fijian counterparts, they
comprised some 46 per cent of the total also had greater income inequality. The
population but accounted for some 47 per 1990/91 Gini coefficient for Indo-Fijians was
cent of those in poverty. Indigenous Fijian 0.51 compared to that of indigenous Fijians
households accounted for 50 per cent of the of 0.39; reflected by the overrepresentation
total household population and some 51 per of Indo-Fijians on the extreme ends of
cent of those in poverty. Thus, Indo-Fijian the income distribution in the data. Indo-
households were slightly over represented Fijian households had incomes below
within the population of poor households. their indigenous Fijian counterparts at
At least in statistical terms, it is not correct each of the deciles until the median when
to suggest that the incidence of poverty the comparisons flip around (Table 2).
was higher amongst indigenous Fijians These conclusions hold for the 2002/03
compared to Indo-Fijians in either of the HIES where rural Indo-Fijians had a per
two recent HIES. capita adult equivalent weekly income of
Table 2 Average weekly household income and average per capita income, 1990/91 (F$)
Average household income Average per capita income
Ten per cent group Indo-Fijian Fijian Indo-Fijian Fijian
1 32.40 38.10 7.10 8.10
2 60.80 67.60 13.60 13.60
3 81.60 89.80 17.80 17.50
4 101.10 107.00 22.10 21.50
5 124.20 126.70 27.30 28.60
6 152.50 147.60 32.60 30.40
7 186.70 175.10 40.90 36.20
8 240.60 217.60 52.50 44.40
9 327.90 288.00 74.80 60.20
10 914.40 537.10 227.60 131.00
Source: United Nations Development Programme, 1997. Fiji Poverty Report, United Nations Development
Programme and Government of the Republic of Fiji Islands, Suva:25.
26
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
F$46.18 compared to the corresponding access to alternate income earning assets
figure for indigenous Fijians of F$49.97 can be a life sentence into poverty and
and that for the nation as whole of F$48.57. with little hope of escape for subsequent
Urban Indo-Fijians, in contrast, had a per generations.
capital adult equivalent weekly income The data shows that while poverty has
of F$70.91 compared to the corresponding increased over time, the incidence of poverty
figure for indigenous Fijians of F$68.93 and is not greater amongst indigenous Fijians
a national average of F$72.43.4 The over compared to Indo-Fijians. The rural sector
representation of Indo-Fijian households has a higher percentage of poor households
on both extremes of the income distribution but there is no discernable difference in the
combined with the limited access to natural incidence of poverty between Indo-Fijians
resources by them as a whole would have and indigenous Fijians. The government’s
left the disadvantaged Indo-Fijians highly policies for poverty reduction, as shown
vulnerable to abject poverty. Access to land later, have claimed otherwise in justifying
is critical for the poorest households. The their affirmative action programs.
HIES data shows that households below
the food poverty line were found to be
disproportionately concentrated in the Redistribution and growth of the
squatter settlements and in urban villages. economy
In terms of changes in the incidence of
poverty, the HIES data show that poverty in Nearly every government handout since
rural areas has risen significantly, increasing independence has been undertaken through
from an estimated 24.3 per cent of the rural the lens of ethnicity. For example, access to
population in 1990/91 to 38.1 per cent by government contracts, business licenses,
2002/03. However, the difference in the scholarships for tertiary study and import
incidence of poverty in rural areas between licenses have all been allocated on the basis
the Fijian (37.3 per cent) and Indian (39.2 per of ethnicity. While access to privileges
cent) populations is only marginal. afforded by the taxpayers on the basis of
There is considerable evidence of the ethnicity dates back to the 1874 when Fiji
presence of a poverty trap in Fiji, a problem was ceded to Britain, redistributive policies
that may have been accentuated by the founded on ethnicity have taken on added
redistributive policies of past governments. significance since independence in 1970
The long-term poor are characterised by a and particularly following the first coup in
lack of secure means of livelihood, limited 1987. The first Governor of Fiji, Sir Arthur
access to skills and education, and restricted Gordon, on taking office on 10 October 1874
access to land and other natural resources. made a decision to minimise the impact of
Female-headed households were found to colonisation on the indigenous population.
be at a particular disadvantage. Children Governor Gordon banned alienation of
from resource-poor schools and families land, established indirect rule via the
have had the greatest tendencies to fail traditional chiefs, and, in the process, created
external examinations and thus drop out a Fijian aristocracy with a vested interest
of the education system. UNDP (1997), for in maintaining a system introduced at
example, notes that at least 30 per cent of all colonisation. In taking this stance, Governor
the children who dropped out of primary Gordon was influenced by the traumatic
schools did so because of financial pressure. experiences of the indigenous people under
A premature end to education with limited colonisation in Africa, Australasia, and the
27
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressPacific Economic Bulletin
Caribbean. However, he had to ensure that Since at independence the population
the economy grew in order to enable Fiji to was divided almost equally between the
fund its public provisions. His immediate indigenous Fijians and Indo-Fijians, the
challenge was in effectively governing a contests in accessing government privileges,
large and dispersed archipelagic colony including political office, was invariably
(Ward 1995; Gillion 1962). between the two major ethnic groups. The
Immigrant labour was brought in to ‘Others’, meaning people of all other races,
grow the economy whilst quarantining comprised less than 10 per cent of the total
the indigenous workers to their traditional population and constituted the more affluent
ways of life in the villages. The majority of Chinese and Europeans, and the relatively
the migrants brought in after colonisation poorer non-indigenous Pacific Islander
were indentured labourers, some 61,000 communities. The divisions between
from India, to work on then European- the two major ethnic groups are most
owned plantations. The ban on alienation evident in the resource sectors and those
of land brought a halt to expansion of entailing access to government handouts,
plantation agriculture; but with Fiji’s including access to senior positions within
rich and abundant tropical land and an the bureaucracy. These privileges entail
administration loathe to draw indigenous what economists call ‘rents’; that is, a pure
labour into the modern economy, this placed surplus accruing to the owner due to the
an even larger premium on a ready supply scarcity value of the product. An élite
of resident labour. Indian migrants and their athlete, for example, commands a premium
progeny filled this void; while they were on scarce talent much like the holder of an
encouraged by the administration to stay exclusive import license. In the case of the
and work land leased from the indigenous athlete, however, the talent is innate while
population, their rights to natural resources in the case of the latter, it is a license to ‘print
remained circumscribed. The indenture money’ conferred on the holder via fiat. The
system ended in 1920 but immigration of distinction is important as the latter can be
agriculturalists and traders, mostly from acquired, a fact that induces competition for
India, continued until independence. 5 its acquisition. Such rent-seeking activity
Thus, the Indo-Fijians as seen through the is directly unproductive as the resources
HIES of 1990/91 and 2002/03 comprise expended in such competition are to the
a heterogeneous mix of descendents of cost of producing other goods and services
indentured labourers, and free migrants who in the economy (Bhagwati 1982). Fractures
came as agriculturalists and traders, with the in the community along racial lines occur
latter group having a monopoly on urban when competition for publicly granted and
business. Furthermore, the effects of the taxpayer-funded privileges intensify—a
quarantining of the indigenous population feature characteristic of post-independence
into villages and away from enterprise, whilst Fiji. The effects of the ‘rent seeking’ have
locking the Indo-Fijians into commerce and spilled over into coups and resulted in
agriculture—mostly on land leased from political turmoil and the exodus of human
the indigenous population—remains in the and financial capital. The cumulative effect
politics of race and income redistribution of the above has been a drag on the rate of
in contemporary Fiji. Ethnic competition in growth of the economy.
politics, however, took an added impetus Privileges created via fiat distort
following independence in 1970. incentives for expending effort into value-
28
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
adding activity. Access to government redistributive policies and one that was
scholarships for secondary and tertiary specifically couched in terms of a poverty-
study, for example, is provided on the basis reduction strategy.
of marks earned in nationally conducted While the redistributive policies owe
external exams. But the bureaucratic their origins to the first colonial regime,
hurdles facing a Fijian trying to gain entry the two coups provided considerable
into university and earn a scholarship are thrust. The Interim Government following
lower than for the other races. Knowing the first coup in 1987 introduced its ‘Nine
this preference, an indigenous Fijian student Point Plan’, which included injections of
has a lesser incentive than their Indo-Fijian funds into Fijian Holdings Limited—a
counterpart to excel at studies. The data company created solely for the benefit of the
supports this conjecture; indigenous Fijian indigenous population—subsidised access
students have, on average, under-performed to finance for the indigenous population
and thus have been unable to fulfill their from state-owned banks, government-
scholarship quotas. Those gaining entry funded business training for indigenous
to university have, on average, opted for Fijians, the establishment of compulsory
the ‘softer’ disciplines knowing that the savings schemes for the indigenous
privileges extend to being offered jobs in the population, and the reservation of half of
bureaucracy on graduation. The response all government contracts and investments in
of the policymakers to the observed under- resource-based activities for the indigenous
performance has been to increase the population. Indigenisation of the public
margins of preferences. The net effect is a service was also aggressively pursued
treadmill on a one way street to waste and while similar expectations were made of
underperformance. the private sector.
In comparison, the Indian student, In 2002, following the third coup,
knowing full well the requirements to over- the Government launched an ambitious
perform, has had the opposite incentives: to policy agenda called the Blueprint for the
excel and take on the more difficult subjects Advancement of Fijians and Rotumans, ‘to
so as to achieve a place at university and a improve the economic and social positions of
job on graduation. The parents have faced the indigenous population in Fiji society’ (Fiji
similar incentives, with Indo-Fijian parents 2002:i). The then Prime Minister, Mr Laisenia
having fewer children but on average Qarase, in launching the 20-Year Development
spending more on their education than their Plan (2001–2020) for the Enhancement of
indigenous Fijian counterparts. The 2002/03 Participation of Indigenous Fijians and Rotumans
HIES shows that indigenous Fijians had an in The Socio-Economic Development of Fiji
average 1.9 children per household while rationalised the discriminatory policies on
the corresponding figure for Indo-Fijians the basis of an: ‘expressed recognition of
was 1.2; similarly, indigenous Fijians spent the paramountcy of Fijian interests and also
F$145 per year on education per child while the principle that the interests of the Fijian
the corresponding figure for Indo-Fijians community are not to be subordinated to
was F$313 (data from Narsey 2006:10, 58). the interests of the other communities’, as
Thus, while the intentions of redistributive per Chapter 2 of the 1997 Fiji Constitution.
policies may have been to bridge the gap He referred to Clause (k), Section 6 the
between the two major ethnic groups, the Chapter which allows for ‘affirmative
effects could have been quite the opposite. action and social justice programs to secure
We next consider the most ambitious of the effective equality of access to opportunities,
29
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressPacific Economic Bulletin
amenities or services for the Fijian and The most damning impact of the
Rotuman people, as well as for other Blueprint is on poverty. The redistributive
communities, for women as well as men, and policies were to be used in micromanaging
for all disadvantaged citizens or groups, are the divvying-up of income at the sector and
based on an allocation of resources broadly subgroup levels. To set some benchmarks, as
acceptable to all communities’. of 2002, indigenous Fijians comprised some
While the legitimacy (or otherwise) of 55 per cent of the population, Indo-Fijians
the Blueprint is an issue for debate by the 41 per cent, and ‘Others’ the remaining 4
experts, I will concentrate on the economic per cent (Table 3). As of 2002, indigenous
impact of the introduced redistributive Fijian household income was 77 per cent
(distortionary) policies on the economy and of the figure for Indo-Fijian households;
on poverty in particular. Prime Minister the target for 2005 was to bring parity
Qarase used the UNDP’s report to justify between the two groups and by 2015 was to
the proposed affirmative action program, achieve parity in each stratum (subgroup)
claiming that of the income. Indigenous Fijians as of
[t]he 1997 United Nations Poverty Report 2002 occupied 51.4 (43.8) per cent of all
revealed that households with the lowest salaried (wage) positions, the target for
level of income were those in rural areas 2005 onwards was to achieve parity in
and outer islands. Again, the majority of each class of employment. Indigenous
these were Fijians (Fiji 2002:i). Fijians accounted for some 61.5 per cent
The report, however, stated categorically of government positions, a figure well
that ‘[p]overty is not concentrated in any above their proportion in total population,
particular sector of Fiji society but is an yet the targets set claim that ‘parity had
under-current across all communities’ been achieved’. Indigenous Fijians had
(UNDP 1997:2). The HIES data for 1990/91 been awarded 42 per cent of public and
and 2002/03 provides evidence that the corporate tenders and 54.5 per cent of
incidence of poverty amongst the indigenous Public Works Department (PWD) tenders;
Fijians was no greater than that for Indo- the target for 2005 was for parity on value.
Fijians, and if anything the converse was Indigenous Fijians had 53.2 per cent of the
probably true. market for rental of office space; the target
The Prime Minister had also argued that was to achieve parity on value. Bus, taxi,
the proposed policies were to pre-empt a and rental car licenses in Fiji are heavily
reoccurrence of future coups, recognising regulated and thus provide significant
that past coups had done ‘irreparable ‘rents’ to the owner and therefore are an
damage’ to race relations and the economy. issue of significant competition. As of 2002
He thus appealed to the citizenry to accept indigenous Fijians accounted for 33 per
the measures proposed in his government’s cent of the total Land Transport Authority
Blueprint. He was wrong on this count (LTA) licenses; the target for 2005 was set
as well. With the benefit of hindsight, the at 40 per cent and for 2010 it was set at 50
Blueprint not only failed to prevent another per cent; and parity in each subgroup was
coup but was probably critical to the ousting to be achieved from 2010 onwards. These
of the PM from office in the very next coup. are a small sample of the targets set in the
The ‘clean up’ campaign that Commodore Blueprint, but they provide a sense of the
Bainimarama had begun as part of his coup extent of redistribution and the details of
was claimed to bring an end to race-based income and wealth redistribution that were
politics in Fiji. being pursued via regulation.
30
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
It is hard to imagine how these targets indigenous Fijian ahead of an equally able
could be achieved in a market economy Indo-Fijian aspirant. Such a process creates
without constraining competition, taxing feedback loops that only encourage greater
production, inducing corruption, and discrimination. Affirmative action programs
incurring waste. Just as one example, the thus have led to a feedback loop with
target for tenders provided by the Public increasing disparities in employment and
Works Department have been achieved, dampened productivity. The cumulative
but at what cost? Half of the tenders can be effects are reduced levels of productivity
reserved for indigenous Fijians, but at the and a lower rate of economic growth than
cost of not getting the most competitive what would have prevailed otherwise.
supplier of the service. Most targets that The impact of affirmative action
have been attained are those where the programs on poverty can be significant
government has had a direct role in its and works via three distinct channels. First,
allocation, but such allocations would have it is the poor who are most dependent on
been made at the cost of meritocracy. Indo- public services such as the provision of
Fijians have lost out in being promoted to basic education, primary healthcare, rural
the senior echelons of the public service: a roads, and crime prevention. Funds used
fact alleged to be due to the discriminatory for affirmative action programs are lost
policies and a reason for many to emigrate. to the provision of the afore-mentioned
It is also common knowledge amongst (given the budget constraint) with the
employers that the risks of losing an Indian poor being robbed of the services that they
employee to emigration are greater than otherwise would have received. Second,
for indigenous Fijians.6 They thus have an discriminatory policies distort incentives
incentive to hire, train, and promote an such that effort is spent on ‘rent-seeking’
Table 3 Affirmative action targets
2002 level Targets
(per cent) 2005 2010 2015 2020
Household incomea 77.0 1:1 Achieve parity in main subgroups
Employment
Salaried 51.4 To achieve parity in each sub-group
Wage employment 43.8 50 To achieve parity in each sub-group
Government 61.5 Parity achieved
Public corporations 65.7 To achieve parity in each sub-group
Public and corporate tenders/contracts 42.0 50 Aim for parity on value
Rental of office space 53.2 Maintain share, aim for parity on value
Public Works Department tenders 54.5 Aim for parity on all tenders
Land Transport Authority licences 33.0 40 50 Aim for parity in
subgroups
a
Compared to Indo-Fijians.
Source: The data is for a small sample of the targets stipulated in Tables 1 and 2 in Fiji, 2002. 20-Year Development
Plan (2001–2020) For the Enhancement of Participation of Indigenous Fijians and Rotumans in the Socio-Economic
Development of Fiji, Parliamentary Paper No. 73, Government Printer, Suva.
31
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressPacific Economic Bulletin
activity rather than on value-adding effort. the rationale of reducing income inequality,
Such distortions spill across the whole have done the very opposite as they have
economy when the public sector is the lead lent themselves to abuse. While the poor
sector, being the largest single employer, miss out on quality public services, the
and thus sets the employment conditions non-favoured élite are able to circumvent
and wages for the formal sector as a whole. the constraints by purchasing these services
There is considerable anecdotal evidence while the favoured group captures the
to suggest that corruption in Fiji has risen proceeds from public handouts.
significantly since the first coup (see Keith- In the meantime, the economy suffered
Reid 2000). Third, the redistributions on several fronts including a rise in ‘rent-
sanctioned via affirmative action programs seeking’ and thus unproductive activity, the
can drive a wedge between the incomes of loss of human and financial capital as those
the poor and the rich. It is only those with with the means and the incentives emigrated,
the means and the information that are the demand for improved governance
able to access the privileges accorded via suffered as the middle class shrank, while the
affirmative action programs. It is unlikely poor bear the brunt of a slowing economy as
that a poor villager, for example, will apply they have no escape from their predicament.
for a taxi license, a government tender, or The economy, consequently, stagnated
for subsidised credit. These privileges have (Chand 2007, this issue). The politics of
been utilised by the élite with the means and redistribution that benefited the élite,
the information to do so. together with a porous border that has
The case of a taxpayer funded bailout to allowed the middle class, Indo-Fijians in
the tune of F$200 million of the National Bank particular, to emigrate has contributed to a
of Fiji (NBF) for bad loans made between
rise in poverty. The military takeover of 2006
1988 and 1992 following the first coup
has been justified as a ‘clean up’ of corruption
illustrates the impact of affirmative action
and the race-based politics. If successful, this
programs on poverty. Justice Shameem,
may see a (slow) turnaround in the economy,
the Director of Public Prosecutions who
the effects of which may trickle down to the
was responsible for the investigations
poor over time. The coup7 could equally
following the exposure of the abuses by
displace one set of ‘rent-seeking’ élite with
the media, noted at a workshop on anti-
another without changing the underlying
corruption organised by the Fiji Chapter of
incentives for and impacts of redistributive
Transparency International that
politics on the poor: which of the above
…what was supposed to be an prevails only time will tell.
affirmative action program to advance This article has argued against poverty-
soft loans to the disadvantaged increasing government redistributions,
indigenous population was in fact a particularly those that distort incentives for
slush fund for the privileged, many of value-adding effort. It is not an argument
whom were not indigenous and some against redistribution for poverty reduction
of whom were cronies of people in but one against poverty-increasing transfers.
authority (Fiji Times, 23 March 2007). The rising incidence of poverty cannot be left
Ratuva (2002) estimates losses of around to go unchecked as its steady march to ‘50/50
F$400 million—equal to some 30 per cent by 2020’ is well underway. Transfers to the
of the 1987 GDP for the nation as a whole— poor directly rather than via the medium
from affirmative action programs. Thus, the of ethnicity would be a better means of
affirmative action policies, provided under addressing this challenge. I suggested this
32
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
change a decade ago (Chand 1997), but the The Governor also outlawed alienation of
politics of such a change then was obviously land, established a council of tribal chiefs
not opportune. It may be now, but only if, in as custodians of the resource, and allowed
Gillion’s terms, everyone in Fiji is equally the leasing of this land by the immigrants
accepted as citizens and Fiji does penance for but with the payments being channeled
the mistakes of the past (Gillion 1962:201). via the chiefs. The last entrenched the
position of chiefs in Fijian society, domestic
politics, and the economy by giving them
Conclusion the imprimatur of the colonial authority.
This was the beginning of the creation of
This article makes four principal claims unequal groups within Fiji: the hierarchy
regarding poverty and redistribution in comprised the Governor at the apex,
post-independence Fiji. First, that poverty followed by the chiefs, with the European
has been on a steady upwards trend since planters somewhere in between, the rest of
independence, rising at a rate of one the indigenous population next, and Indian
percentage point per year. One in every migrant workers last.8
four in the population were unable to afford Third, income and wealth redistribution
the minimum requirements for living in on the basis of ethnicity has lent itself to
1990/91—this figure had risen to one in capture by the élite and particularly so in
every three by the next HIES of 2002/03; the heightened climate of racial segregation
and, extrapolating on past trends, one in following the first coup. Thus, while the
every two will be in poverty by the year 1997 Constitution highlighted the poverty-
2020. The last would bring new meaning to reduction rational for redistribution, politics
the Blueprint of the Qarase Government that consolidated a system of redistribution
had espoused a vision of half of the economic that was poverty-increasing. Household-
wealth of the country being in the hands of level data show that poverty is prevalent
the indigenous population by the year 2020, in all communities, and that indigenous
thus their slogan of ‘50/50 by 2020’. Fijians are no worse off than Indo-Fijians
Second, affirmative action programs, in terms of the proportion of the population
of which the blueprint was their last in poverty. Post-coup administrations
manifestation, have been adopted since in Fiji, however, have argued otherwise;
the beginning of colonisation when the claiming that indigenous Fijians face
first resident Governor, Sir Arthur Gordon, greater disadvantage than Indo-Fijians.
experimented with ‘non-transformative’ Surprisingly, these claims have principally
colonisation in this archipelagic colony been based on the Fiji Poverty Report (UNDP
(Veracini 2007). Indigenous Fijians were 1997). These claims have been used to
to be quarantined from modern commerce rationalise affirmative action programs
and particularly from the likely abuse entailing redistribution of taxpayer funded
from the resident European planters. handouts to the indigenous population. This
Migrant labour, the majority from India, ethnicity-based redistribution has been anti-
was brought in to save the indigenous poor; an observation consistent with earlier
Fijians from this predicament and grow the findings of Cameron (1983) and Stavenuiter
economy. Governor Gordon supplemented (1983). It has also been a drag on the growth
the meager resources he had at his disposal of the economy, and a major source of
to govern the colony by drawing in the corruption and abuse of public office. The
tribal chiefs as part of his administration. discriminatory practices, moreover, have
33
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressPacific Economic Bulletin
induced emigration and capital flight, thus (1962) for an account of the separatist
leaving those without ‘wings’, the immobile polices of the colonial administration, thus
poor in the main, face the brunt of the costs giving legitimacy for racial segregation and
of discrimination. ensuing political and social ramifications of
immigration for Fiji since.
Last, the economy has largely stagnated 6
Frank Bainimarama has claimed that ‘[r]acist
since the first coup. Per capita income policies really have been the main cause of
since 1987 has grown at less than 1.5 per brain drain in the public sector since the 1987
cent per annum. Investment rates have coups [and] racism in the private sector also
plummeted following each coup while was the major cause of the brain drain there’
annual emigration of skilled personnel has (reported by Fijilive, 1 April 2007).
doubled since the first. The poor usually 7
This was a military takeover without the
have only their labour to sell and often abrogation of the constitution or a revolution,
at unskilled wage rates. Average daily thus it may not amount to being a coup in the
strict sense of this term.
real wages fell in the first two decades of 8
Remnants of this hierarchy and the Gordon
independence while income inequality experiment remain in contemporary Fiji.
increased in this period: the two together I cannot help but ask if the indentured,
imply that the poor must have lost ground often illiterate and uninformed, Indian
in terms of per capita income. labourer was recompense for the atrocities
The 2006 coup was rationalised as being of European colonialism elsewhere.
undertaken to ‘clean up’ the government of
corruption and to bring race-based politics References
to an end. The intentions are laudable
but only time will tell if the effects will be Abbott, D.F., 2007. Fiji analysis of the
consistent with the intentions. 2002/03 household income and
expenditure surveys: estimation
Notes of basic needs poverty lines and
incidence of poverty in Fiji, Draft Final
1
This claim is subject to all available data on Report, prepared for United Nations
this issue. Development Programme, Suva
2
Fiji has had five-yearly national development (unpublished).
plans dating from independence in 1970
Asia Pacific Migration Research Network,
to 1991, each of which emphasised broad-
based growth and improved access to basic n.d. Migration issues in the Asia
services to the disadvantaged and the poor. Pacific: issues paper from Fiji, Asia
The government created a Poverty Taskforce Pacific Migration Research Network
in 1991 to recommend polices to reduce the Suva. Available from http://www.
burgeoning levels of poverty. unesco.org
3
Income inequality in Fiji, as measured by Bhagwati, J., 1982. ‘Directly unproductive,
the Gini coefficient (which ranges from 0 for profit-seeking DUP activities’, Journal
perfect equality and 1 for perfect inequality),
of Political Economy, 90:988–1,002.
rose between 1977 and 1991 from 0.43 to 0.49
(UNDP 1994). Cameron, J., 1983. The extent and
4
Note that ‘Others’ have the highest income, structure of poverty in Fiji and a possible
dragging the national average up. The figures Government Anti-Poverty Strategy,
for 2002/03 are from Narsey (2006). Discussion Paper No. 145, University
5
See Lal (2004) on the origins of Fijian Indians of East Anglia, Norwich.
and an account of the experiences of the Chand, S., 1997. Ethnic conflict, income
indentured labourers in Fiji, and Gillion
inequality, and growth in independent
34
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific Presspoverty and redistributive politics in post-independence Fiji
Fiji, State Society and Governance Ratuva, S., 2002. ‘Economic nationalism
Discussion Paper No. 97–6, The and communal consolidation:
Australian National University, economic affirmative action in Fiji,
Canberra. 1987–2002’, Pacific Economic Bulletin,
——, 2007. ‘Swim or sink: the predicament 17(1):130–37.
of the Fiji economy‘, Pacific Economic Stavenuiter, S.,1983. Income distribution in
Bulletin, 22(2):1–21. Fiji: an analysis of its various dimensions,
Chandra, R. and Chetty, K., 1996. with implications for future employment,
Report on emigration, Asia Pacific basic needs and income policies, WEP
Migration Research Network, Suva Research Working Paper, International
(unpublished). Labour Organization, Geneva.
Fiji, 1993. Opportunities for Growth: Policies United Nations Development Programme
and Strategies for Fiji in the Medium (UNDP), 1994. Pacific Human
Term, Parliamentary Paper No. 2, Development Report, UNDP, Suva.
Government Printer, Suva. ——, 1996. Human Development Report
——, 1996a. Supplement to 1996 Budget 1996: economic growth and human
Address (unpublished). development, UNDP, New York.
——, 1996b. Mission 2010: Mission ——, 1997. Fiji Poverty Report, UNDP and
statement to the Year 2010 by the Government of the Republic of Fiji
Government of the Republic of Fiji Islands, Suva. Available from http://
(draft) (unpublished). www.undp.org.fj
——, 2002. 20-Year Development Plan ——, 2006. Human Development Report
(2001–2020) For the Enhancement of 2006. Beyond Scarcity: power, poverty and
Participation of Indigenous Fijians the global water crisis, UNDP, New York.
and Rotumans in the Socio–Economic Available online at http://hdr.undp.
Development of Fiji, Parliamentary org (accessed on 14 March 2007).
Paper No. 73, Government Printer, Veracini, L., 2007. ‘Emphatically not a
Suva. white man’s colony: settler colonialism
Gillion, K.L., 1962. Fiji’s Indian Migrants. and the construction of colonial Fiji’,
A history to the end of indenture in 1920, The Australian National University,
Oxford University Press, Melbourne. Canberra (unpublished).
Kanbur, R., 1984. ‘Inequality, poverty and Ward, R.G., 1995. ‘Land, law and custom:
development with an application to diverging realities in Fiji’, in R.G. Ward
Fiji’, Seminar Paper No. 365, University and E. Kingdon (eds), Land, Custom and
of Essex, Colchester. Practice in the South Pacific, Cambridge
Keith-Reid, R., 2000. ‘Pacific Corruption: University Press, Cambridge:198–249.
it’s out there. But can the rot be
stopped?’, Island Business (online), 15 Acknowledgments
May. Helpful comments from Ron Duncan,
Lal, B.V., 2004. ‘Girmitiyas: the origins Azmat Gani, Steve Pollard, Biman Prasad
of the Fiji Indians’, Fiji Institute of and Lorenzo Veracini on an earlier draft
Applied Studies, Lautoka. of this paper is gratefully acknowledged.
Narsey, W., 2006. Report on the 2002–03 A version of this paper is to appear in an
Household Income and Expenditure edited volume by Brij Lal on a 20 year
Survey, Vanuavou Publications, Suva. retrospective on the first Fiji coup.
35
Pacific Economic Bulletin Volume 22 Number 2 2007 © Asia Pacific PressYou can also read
